Subscribe to DSC Newsletter

Motivation

Education for years has been one model that fits for all. A fixed curriculum developed based on experience is delivered to a group of students with the hope of it getting across to all the students with the same effectiveness.  Then each student is graded based on the basis of his understanding of that curriculum. 

Now imagine a classroom where its not just the student that is learning about the world but where the classroom is also learning about the student. Each day is a new learning experience for both, the classroom and the student. The adaptive classroom creates and delivers content specific to the needs of each student individually based on what it learnt about that student.

Premise

In recent years with the use of the Big Data and machine learning, exciting opportunities have been created in order to enhance a student’s learning experience. Few such fields are [1]

  • Learning Analytics and Student Modeling
  • Content Analytics and Text Mining
  • Teaching policy optimization and personalization
  • Automatic and Peer Grading
  • The cognitive science of learning
  • Active learning and experimental design
  • Data analysis for emerging educational platforms

In this paper the focus would be on discussing the advancements in and the use of Learning Analytics.

Learning Analytics

Learning Analytics is defined as “the measurement, collection, analysis, and reporting of the data about the learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs” [2]. All of this has been made possible by making use of the age of Big Data.

Making sense of Big Data and using it to improve the learning process is a challenge. But is can be overcome by making use of Data Mining Techniques such as Regression, Nearest Neighbor, Clustering, and Classification.

How can Data Mining techniques be used in the Learning Analytics in the field of Education?

Data Mining techniques in Learning Analytics can be used to attain the following outcomes [3] 

  • Prediction of Student’s performance - Students’ performance can be predicted based on their reaction to the teaching methodologies and pedagogy used under certain environment. Use of the data from behavioral analysis, written and oral tests, and their participation data in class.
  • Risk of Attrition - Based on the students’ social and economical condition, interest in the subject, career aspiration, behavior, etc. the risk of a student dropping out of class, school or college can be measured. Corrective measures can be taken at the early stages itself to prevent high attrition rate by creating a conducive learning environment for each student.
  • Understanding of the student - Reporting on Big Data can get difficult as it continues to grow every moment. For a teacher to know about the performance, behavior, needs are of utmost importance. Use of Data Visualization techniques can help a facilitator to understand the situation better and thus improve the quality of education pedagogy.
  • Intelligent Feedback - Based on the student’s performance, an immediate and intelligent feedback can be sent to both to the student and the facilitator to help them understand the learning needs better. A student will be able to take the corrective measures in real time and thus facilitating to better understanding and retention of the subject.
  • Recommendation of Courses - New courses can be recommended to the students based on their interest. These interests can be identified directly by an interactive conversation with the student or indirectly by analyzing historical data about that student and also by making use of the data stored about other students with similar interest and abilities.
  • Student Skills Estimation - Data Mining can be used here in identifying a student’s skills in two ways. First, by making use of the data collected through assessments and other techniques about that student while in the program. Second, by estimating the skills of the student based on assessments taken at the beginning of the program and comparing it with the historical data. This will help in designing the most suitable curriculum for that student not only by making use of strengths but also by keeping into account his interest and working of on his weaknesses to help him build a career in the field of his choice.
  • Grouping and collaboration of students - With the use of Big data, different types of groups can be created that will assist in enhancing student learning from his peers.
  • Planning and Scheduling - Intelligent study plans can be created for the student to help him learn at his pace. Curriculum can be designed in ways that it delivers the right content at the right stage so that topics can complement each other and make a continuous story for that student. 

Levels of Learning Analytics

There are three major levels of learning Analytics [4]

  • Macro-Level Analytics - This level of analytics talks about comparisons, learnings, studies among institutions. This can help in assist in learning from the best practices, standardizing practices across institutions, states, countries, etc. Over the period of time Macro level learning analytics will become more granular in nature when data from Meso and Micro level analytics will also be fed into the system.
  • Meso-Level Analytics - This learning level functions at the level of a single institution. Use of Business Intelligence process that are already present in the market to improve the overall functionality and feed this information into the Macro – Level Analytics System.
  • Micro-Level Analytics - At this level data about the processes tracking the progress of an individual learner is stored and analyzed. This data is used by the learner himself and by the the people that are involved in the development of the learner or student.

Place of Pedagogy in Learning Analytics

It is unclear at the moment where the Pedagogy really stands in the Learning Analytics but few points can be made based on the advancements made recently in terms of collection and availability of data. In my opinion, pedagogical methodologies shouldn’t be made a part of the Learning Analytics but should be served by it. Outcomes from Learning Analytics should feed into creation and improvement of teaching pedagogy.

Examples of Learning Analytics [4] 

  • Analytical Dashboards in online learning platforms.
  • Predictive Analytics – From a learner’s static and dynamic data it be can be predicted the trajectory he is on. With this knowledge timely interventions can be made to assist the student achieve the most from his learning experience.
  • Adaptive Learning Analytics – Building a model that understands the learning type and progression of a student and helps create a curriculum that is suited to the learning needs of each individual student.
  • Social Network Analytics – Analysis of the interpersonal relations of the candidates or students to analyze their skills towards creating, maintaining, and using those relationships.
  • Discourse Analytics – Developing systems that can assess the quality of text or argument.

Common Software in use for Learning Analytics

  1. Student Success System [5]
  2. SAP-HANA [6]
  3. Blackboard [7]
  4. SNAPP [8]
  5. LOCO-Analyst [9]

Limitations

There are quite a few technological and ethical limitations that need to be overcome for a successful implementation of Learning Analytics.

In lesser developed countries where access to internet is not available everywhere, implementing a real time machine learning system will be a challenge.

Availability of the required hardware – computers, tablets, smartphones – in developing nations or nations with high populations where in a class there can be more than 50 students is extremely resource heavy and difficult to achieve without right backing of the project.

Collection and storage of data about the student and his social structure can be termed as unethical by quite a few.

Extinction of facilitator based learning. It can be argued that the touch of a human in learning is of most importance and as the Machine Learning systems will continue to grow, the dependency on them for learning will increase.

Data Ownership.

Conclusion

As in every field use of the data and Analytics is gaining momentum, Learning Analytics in Education is also growing. It is the future of learning. The availability of data has given us the power to improve the learning experience of each child and creating a better future for him and for the entire society. With more and more bigger organizations such as IBM and SAP taking deep interests in Machine Learning for Education a future where education will just not be books and examinations is not far.

Having said that, it should also be brought to the attention of the reader that it is too early to correctly determine the potential and role of Learning Analytics in the education system.

In the end any system, be it based on a facilitator or based on machine learning, is developed to improve the learning experience of a child. After all, what is education without learning.

References

[1] http://dsp.rice.edu/ML4Ed_ICML2015

[2] Long & Siemens, 2011, page 32

[3] Manolis Mavrikis, Patricia Charlton and Demetra Katsifli, “The Potential of Learning Analytics and Big Data”, http://www.ariadne.ac.uk/issue71/charlton-etal.

[4] http://iite.unesco.org/pics/publications/en/files/3214711.pdf

[5] http://www.brightspace.com/solutions/higher-education/advanced-anal...

[6] http://hana.sap.com/abouthana.html

[7] http://www.blackboard.com/

[8] http://www.snappvis.org/

[9] http://jelenajovanovic.net/LOCO-Analyst/

Age of the Geeks!

Views: 3669

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Kening Ren on November 5, 2015 at 3:11pm
Thank you for sharing this, especially those software tools.
Comment by Sione Palu on November 5, 2015 at 9:32am

Quote "Education for years has been one model that fits for all. A fixed curriculum developed based on experience is delivered to a group of students with the hope of it getting across to all the students with the same effectiveness.  Then each student is graded based on the basis of his understanding of that curriculum."

I agree here,  education is too crispy & rigid but not malleable & flexible. Teachers & Experts insisted that their way is right and anything outside the one size fits all systematic thinking must be wrong.

Videos

  • Add Videos
  • View All

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service